TY - GEN
T1 - Design and Implementation of an IoT-Based System for Monitoring Endangered Beaches
AU - Rivadeneira, Franco
AU - Arauco, Bonnie
AU - Teran, Ariana
AU - Flores-Oscanoa, Jesus
AU - Casano, Celso De La Cruz
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Beach pollution is an escalating global issue, posing a significant threat to marine ecosystems and human health. Plastic waste, originating from commercial activities, fishing, and improper waste management, is prevalent on beaches worldwide. This debris is transported by rivers and waterways, leading to widespread contamination. Additionally, climate change and pollution are expected to result in the loss of many sandy beaches, altering ecosystems and diminishing natural recreational areas. For this reason, an IoT-based system has been developed that, using low-cost autonomous cleaning robots on different beaches, can collect data and detect garbage using the new YOLO V10 architecture. As a result, the system was successfully implemented to store and alert users when a beach has a high level of pollution. It efficiently collects and saves data from multiple beaches in a safe NoSQL Database using multiple robots simultaneously. Additionally, the highest performance predictive model achieved an accuracy of 77.69%, a recall of 71.63%, and an F1 score of 74.53%.
AB - Beach pollution is an escalating global issue, posing a significant threat to marine ecosystems and human health. Plastic waste, originating from commercial activities, fishing, and improper waste management, is prevalent on beaches worldwide. This debris is transported by rivers and waterways, leading to widespread contamination. Additionally, climate change and pollution are expected to result in the loss of many sandy beaches, altering ecosystems and diminishing natural recreational areas. For this reason, an IoT-based system has been developed that, using low-cost autonomous cleaning robots on different beaches, can collect data and detect garbage using the new YOLO V10 architecture. As a result, the system was successfully implemented to store and alert users when a beach has a high level of pollution. It efficiently collects and saves data from multiple beaches in a safe NoSQL Database using multiple robots simultaneously. Additionally, the highest performance predictive model achieved an accuracy of 77.69%, a recall of 71.63%, and an F1 score of 74.53%.
KW - Artificial Intelligence
KW - Cloud Services
KW - Computer Vision
KW - IoT
UR - http://www.scopus.com/inward/record.url?scp=85217868959&partnerID=8YFLogxK
U2 - 10.1109/IHTC61819.2024.10855065
DO - 10.1109/IHTC61819.2024.10855065
M3 - Conference contribution
AN - SCOPUS:85217868959
T3 - 2024 7th IEEE International Humanitarian Technologies Conference, IHTC 2024
BT - 2024 7th IEEE International Humanitarian Technologies Conference, IHTC 2024
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 7th IEEE International Humanitarian Technologies Conference, IHTC 2024
Y2 - 27 November 2024 through 30 November 2024
ER -